O-glycoside orientation is an essential aspect of base J recognition by the kinetoplastid DNA-binding protein JBP1

Article abstract of DOI:10.1002/anie.200604635

A binding edge? Comparison of dissociation constants and molecular-dynamics snapshots of a panel of synthetic telomeric double-stranded DNA sequences containing HMdU O-glycosides has revealed that the DNA-binding protein of kinetoplastids, such as Leishma

Full text of DOI:10.1002/anie.200604635

Angewandte
Chemie
DOI: 10.1002/anie.200604635
Molecular Recognition
O-Glycoside Orientation Is an Essential Aspect of Base J Recognition
by the Kinetoplastid DNA-Binding Protein JBP1**
Rajesh K. Grover, Stephanie J. K. Pond, Qizhi Cui, Prem Subramaniam, David A. Case,
David P. Millar, and Paul Wentworth, Jr.*
Leishmaniasis (cutaneous or visceral, Leishmania sp.), Afri-
can trypanosomiasis (sleeping sickness, Trypanosoma brucei),
and American trypanosomiasis (Chagasꢀ disease, T. cruzi) are
devastating diseases in the developing world. Present phar-
maceuticals against these parasites are limited in terms of
effectiveness, increasing drug resistance, and inherent drug
toxicity. Therefore, there is a clear need for elucidation of
parasite-specific biological targets that would be amenable to
new therapeutic approaches.
In this regard, Borst and co-workers[1–3] made a landmark
contribution when they discovered that DNA from kineto-
plastida and Euglena,[4] but not other eukaryotes, contains a
modified nucleobase, 5-(b-d-glucopyranosyloxymethyl)-2’-
deoxyuridine (1a), called nucleoside dJ or base J (Figure 1).
Base J is found in telomeric repeats of all kinetoplastida[1] and
in inactive telomeric variant surface glycoprotein (VSG) gene
expression sites in T. brucei.[5–9] Borst and co-workers[10]
subsequently found that extracts of several kinetoplastid
flagellates contain a DNA-binding protein that specifically
binds to dJ-containing duplexDNA. The role of this dJ-
binding protein 1 (JBP1) is yet to be clearly defined, but JBP1
[*] Dr. R. K. Grover, Prof. P. Wentworth, Jr.
Department of Chemistry and
The Skaggs Institute for Chemical Biology
The Scripps Research Institute
10550 North Torrey Pines Road, La Jolla, CA 92037 (USA)
Fax: (+1)858-784-7835
E-mail: paulw@scripps.edu
Prof. P. Wentworth, Jr.
The Scripps-Oxford Laboratory
Oxford Glycobiology Institute
Department of Biochemistry
University of Oxford
South Parks Road, Oxford, OX13QU (UK)
noethyl-N,N’-diisopropyl)phosphoramidite derivatives (2a–h) for incor-
Dr. S. J. K. Pond, Dr. Q. Cui, Prof. D. A. Case, Prof. D. P. Millar
poration into the telomeric 16-mer oligonucleotide sequence (tel-J*)
Department of Molecular Biology
Figure 1. Base dJ (1a) and HMdU O-glycoside analogues (1b–h) were
synthesized as their peracetylated 5’-O-dimethoxytrityl-3’-O-(2-cya-
shown. J* is the locus of incorporation of 1a–i. DMT=dimethoxytrityl.
The Scripps Research Institute
10550 North Torrey Pines Road, La Jolla, CA 92037 (USA)
gene knock-outs of Leishmania tarentolae have revealed that
JBP1 is essential for its survival,[11] hinting that inhibition of
JBP1/dJ-DNA binding may offer merit as a new therapeutic
strategy against Leishmania.
The design of molecules that may disrupt this interaction
would clearly benefit from structural information, at atomic
resolution, regarding the nature of this relatively rare
molecular interaction in biological systems—that between a
glycosylated oligonucleotide and a protein. However, at
present, there is no X-ray structural analysis of either JBP1
alone or in complexwith dJ-containing DNA to guide library
Dr. P. Subramaniam
Department of Infectology
Scripps Research Institute—Florida
5353 Parkside Drive RF-2, Jupiter, FL 33458 (USA)
[**] This work was supported by the Skaggs Institute for Chemical
Biology (P.W.). We thank Dr. P. Ordoukhanian of the TSRI Nucleic
Acid Core and E. J. C. van der Schans for oligonucleotide synthesis
and purification. P.W. and R.K.G. thank M. Puga for assistance
during the synthesis of 1a–h. P.S. thanks Prof. C. Weissmann for his
assistance and support.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 2839 –2843
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2839

Communications
design. And, while the amino acid sequence of JBPs from
several kinetoplastids are known and all share significant
degrees of homology (43–67%),[10] the unique nature of these
sequences has, thus far, meant that homology modeling has
proven unsuccessful.
Previous studies by Sabatini and co-workers have
delineated the oligonucleotide requirements for the binding
of JBP1 to dJ-containing DNA.[12,13] These studies reveal that
JBP1 binds only to double-stranded dJ-containing DNA and
does so optimally when the dJ base is flanked by at least five
other nucleotide residues in a telomeric sequence (KD
ꢀ 40 nm, measured using gel shift assays). DNA footprinting
techniques further reveal that JBP1 does not make any
sequence-specific contacts with the nucleosides adjacent to
base J and that the only critical contacts between JBP1 and
dJ-DNA occur on the dJ-containing oligonucleotide. These
contacts consist of thymine-dependent major- and minor-
groove contacts at base dJ and a sequence-independent
major-groove contact at the nucleotide 5’ to J (termed the J-
1 nucleotide). When dJ is replaced with dT, the nucleobase
considered to be the biosynthetic precursor of genomic dJ,
Sabatini et al.[12] showed that JBP1 binding is reduced to a
level undetectable by gel shift assay.
Herein, we report the first in-depth analysis of the
molecular recognition between the O-linked glycoside com-
ponent of dJ in telomeric dJ-containing double-stranded (ds)-
DNA and JBP1. Comparison between the molecular dynam-
ics (MD) snapshots and the DDG values of the binding of
JBP1 to the duplextel-J* oligonucleotides 1a–h reveals that
JBP1 binding to dJ-containing oligonucleotides occurs pref-
erentially when the b-d-glucopyranosyl moiety adopts a
conformation within the major groove wherein the C2 and
C3 hydroxy groups of the glucoside form hydrogen bonds
with the nonbridging pro-R phosphoryl oxygen of the J-1
nucleotideꢀs phosphate group. If this orientation is perturbed
even slightly, then JBP1 binding affinity drops to the level of
that when dJ is replaced by dT (100-fold, ꢀ 11 kJmolÀ1).
The N3-unprotected phosphoramidite analogues 2b–h
(Figure 1) were prepared based upon an original method for
the N3-unprotected phosphoramidite 2a, developed by van
Boom and co-workers (Scheme 1).[14–16] Thus, bis-silylation of
dT (1i), followed by radical allylic bromination gave allylic
bromide 3 in good yield (67% over two steps). Nucleophilic
displacement of the crude alkyl bromide 3 with cesium
methoxyacetate in DMF followed by saponification with
K2CO3 in methanol gave bis-silylated 5’-hydroxmethyl deoxy-
uridine (HMdU) 4 in 72% yield (a considerable improvement
over the published yield of 51%).[15] The critical glycosylation
of 4 with the corresponding sugars required significant
optimization. Optimal glycosylation of 4, coupled with b-
anomeric selectivity, was achieved by means of a Schmidt
condensation[17] in the presence of trimethylsilyltriflate
(TMSOTf) at À258C, with the sugars as their corresponding
perbenzoylated a-trichloroacetimidate esters. These condi-
tions not only served to minimize formation of the corre-
sponding and unwanted a anomer but also prevented HMdU
dimerization.
Scheme 1. Synthesis of glycosylated phosphoramidites 2a–h.
a) 1. TBDMS-Cl, imidazole, DMF; 2. NBS, AIBN, C6H6, 67% over two
steps; b) 1. cesium methoxyacetate, DMF; 2. K2CO3, CH3OH, 71%
over two steps; c) 1. perbenzoylated a-trichloroacetimidate ester
sugars, TMSOTf, À258C, DCE; 2. NaOMe, CH3OH; 3. Ac2O, Pyr, 94%
over two steps; 4. Et3N·3HF, Pyr; 5. DMTCl, Pyr, RT; 6. P(OCE)-
(NiPr2)Cl/Et3N. Note 2h is prepared by treatment of 3 with NaOMe,
95% followed by c) steps 5 and 6. AIBN=azobisisobutyronitrile,
DCE=1,2-dichloroethane, DMF=dimethylformamide, NBS=N-bro-
mosuccinimide, Pyr=pyridine, TBDMS=tert-butyldimethylsilyl.
analogues, for the purposes of oligonucleotide synthesis,[15]
was then achieved by treatment with NaOMe/MeOH fol-
lowed by acetic anhydride to give the peracetylated sugar-
HMdU b-glycosides in high yield (94% over two steps).
Desilylation, followed by protection at 5’-O with
a
dimethoxytrityl (DMT) group and conversion into the 3’-O-
(2-cyanoethyl)(NiPr2)phosphoramidite gave the nucleoside
building blocks 2a–g required for DNA synthesis.
The modified base, J*, was then introduced into the
telomeric oligonucleotide sequence, 5’-CCCTAACCC-
J*AACCCT-3’, using conventional phosphoramidite solid-
phase DNA synthesis methods. All the telomeric J*-contain-
ing DNA sequences were purified using denaturing poly-
(acrylamide) gel electrophoresis (PAGE) and purity (> 95%)
was confirmed by reversed-phase (RP)-HPLC and denaturing
PAGE.
The
complementary
sequence,
5’-NH2-
(CH2)6AGGGTTAGGGTTAGGG-3’, was prepared and
purified as described above, and then coupled with Cy5-N-
hydroxysuccinimidyl ester and purified by RP-HPLC. DNA
annealing between the J*-containing oligonucleotides and the
Cy5-labeled counterstrand was performed by heating a 1:2
mixture of the Cy5-labeled strand and the J*-base strand to
958C for 5 min in tris(hydroxymethyl)aminomethane hydro-
chloride (Tris-HCl) buffer (20 mm, pH 7.9) containing KCl
(50 mm) followed by slow cooling to 48C. Duplexformation
was verified by native 20% PAGE.
Recombinant JBP1 from the kinetoplastid Crithidia
fasciculata containing a His10 tag was expressed in E. coli
and purified by metal ion affinity column as previously
described[12] and shown to be over 98% pure by sodium
dodecylsulfate (SDS) gel electrophoresis. Prior to the binding
assays, the protein was dialyzed into binding buffer, 2-[4-(2-
hydroxyethyl)-1-piperazinyl]ethanesulfonic acid (HEPES;
35 mm, pH 7.9), ethylenediaminetetraacetic acid (EDTA;
1 mm), KCl (50 mm), MgCl2 (5 mm), and 1,4-dithiothreitol
(DTT; 1 mm). Isotherms for the binding of recombinant JBP1
to the tel-J*-oligos (20 nm) were then measured by fluores-
cence anisotropy (excitation: 590 nm, emission: 680 nm) at
208C. The data were fit to a model for 1:1 association[18] to
determine the dissociation constant, KD (Table 1).
Conversion of the resultant perbenzoylated HMdU O-
glycosides into their more ammonia-labile peracetylated
The measured KD value for the binding of JBP1 to tel-J*
(where J* is 1a) is KD(1a) = (13 Æ 8) nm, and when dJ (1a) is
2840 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 2839 –2843

Angewandte
Chemie
replaced with dT (1i), the measured KD drops 105.4-fold
[KD(1i) = (1370 Æ 15) nm]. This is equivalent to a free energy
difference (DDG) between dJ and dT of 11.35 kJmolÀ1. The
measured binding constants for JBP1 and tel-J* oligos
containing the dJ analogues 1b–h all lie between the
measured values for 1a and 1i. What is immediately striking,
is that epimeric modifications of single hydroxy groups
the non-bridging pro-R phosphoryl oxygen of the adjacent 5’-
nucleotide residue (cytidine-9) (Figure 2b). The orientation
of the pyranosyl ring is roughly parallel to the axis of the
duplexand perpendicular to the plane of the major groove.
No other significant contacts are made between the glycoside
and the DNA, and the hydroxymethyl group on C5 of the
pyranosyl ring is seemingly free to rotate within the major
groove.
Table 1: Binding data for JBP1 (Crithidia fasciculata) todsDNA containing
base dJ (1a), analogues 1b–h, and dT (1i).
Entry
J*
KD[a] [nm]
f[b]
DDG[c] [kJmolÀ1
]
1
2
3
4
5
6
7
8
9
Glc (1a)
Man (1b)
All (1c)
Gal (1d)
4dGlc (1e)
Xyl (1 f)
Rib (1g)
Me (1h)
dT (1i)
13Æ8
1330Æ50
1220Æ44
1240Æ44
270Æ10
57Æ8
1.0
0.00
11.27
11.06
11.10
7.39
102.3
93.8
95.4
20.8
4.4
88.5
91.5
105.4
3.61
1150Æ36
1190Æ43
1370Æ15
10.92
11.00
11.35
[a] Binding titrations for each modified base were performed multiple
times, using different batches of protein, and Kd values are reported as
the mean (Æstandard deviation) of at least three measurements.
[b] f=KD(1b–h)/KD(1a). [c] DDG=ÀRTlnf.
around the pyranosyl core of dJ can cause a reduction in
JBP1 binding that is equivalent to that observed for the dJ-to-
dT modification (see above). Thus, the stepwise equatorial-
to-axial epimerization of hydroxy groups of C2, C3, and C4—
replacing the Glc of dJ (1a) with Man (1b), All (1c), and Gal
(1d)—results in a loss in binding affinity for JBP1 of 102.3-,
93.8-, and 95.4-fold, respectively (Table 1). Other structural
modifications that cause significant reductions in JBP1 bind-
ing are the replacement of the O-glycoside of dJ with Me (1h)
or Rib (1g); these changes cause a 91.5- and 88.5-fold drop in
JBP1 binding affinity, respectively. There are structural
modifications to dJ that do not have such a large impact on
JBP1 binding. Deletion of the hydroxy group on C4 or the
hydroxymethyl group on C5 of the b-d-glucopyranosyl ring to
give the 4dGlc (1e) and Xyl (1 f) analogues results in a
reduction in JBP1 binding affinity of only 20.8- and 4.4-fold,
respectively.
To gain insight into how structural modifications to the
HMdU O-linked glycosides 1a–h may be impacting JBP1
binding by means of changes in the conformation adopted by
the glycoside within the major groove of tel-J*, we studied the
conformational dynamics of the tel-J* ds-DNA oligonucleo-
tides. A 5- to 10-ns molecular dynamics (MD) simulation was
performed for each of the oligonucleotide tel-J* sequences
containing 1a–h. The AMBER 8[19] suite of programs with an
all-atom force field (ff99)[20,21] and the GLYCAM carbohy-
drate force field (glycam04)[22–24] were used to build the initial
duplexstructures of hydrated tel-J* DNA sequences and to
run all MD simulations. The MD simulation of the native dJ
(1a) containing tel-J* sequence shows that the b-d-glucopyr-
anosyl ring adopts a chair conformation and extends into the
major groove (Figure 2a). The glucose is anchored to the
DNA backbone by two hydrogen bonds between the equa-
torial hydroxy groups on C2 and C3 of the pyranosyl ring and
Figure 2. MD snapshot of tel-J* (surface rendering) showing the b-d-
glucopyranosyl ring of 1a (stick representation) occupying the major
groove. a) Overall view, showing the complete telomeric sequence;
b) zoom view, focusing on the orientation of the glucoside within the
major groove. Note the two hydrogen bonds (shown as dotted lines)
between the hydroxy groups of C2 and C3 of the glucopyranoside and
the nonbridging pro-R phosphoryl oxygen of the J-1 nucleotide. The
nucleotides J and J-1 are colored gray with the exception of the DNA-
backbone phosphorus atoms of the J and J-1 nucleotides, which are
orange, and the phosphoryl oxygen atoms, which are red. O-glycoside
oxygen atoms (red), carbon atoms (yellow), hydrogen atoms (white).
Hydrogen bonds are shown as dotted lines.
Inspection of the MD simulations of the tel-J* double-
stranded DNA analogues 1e and 1 f reveals that the hydroxy
groups of C2 and C3 of the pyranosyl ring form hydrogen
bonds with the J-1 pro-R phosphoryl oxygen and the glycoside
is locked in a conformation almost identical to that for the
native dJ (1a) (Figure 3a,e,f). In contrast, for analogues 1b,
1c, 1d, and 1g these hydrogen bonds between the hydroxy
groups of C2 and C3 and the DNA backbone are disrupted
and the conformation of the respective glycosides in the major
groove differs considerably from that observed for the dJ-
containing oligonucleotide (Figure 3b–d). Man(1b)-contain-
ing tel-J* forms no apparent hydrogen bonds with the tel-J*
DNA backbone (Figure 3b). The All (1c) O-glycoside forms
a hydrogen bond between the equatorial hydroxy group on
C4 and the pro-R phosphoryl oxygen of the J-1 nucleotide,
with the axial hydroxy group on C3 being rotated away from
the DNA backbone (Figure 3c). In the case of Gal (1d), the
axial hydroxy group on C4 makes a hydrogen bond with the J-
Angew. Chem. Int. Ed. 2007, 46, 2839 –2843
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2841

Communications
once the orientation optimal for binding of these hydroxy
groups is established by locking of the pyranosyl ring.
If the HMdU O-glycoside cannot adopt this edge-on
conformation, the loss in binding energy of the O-glycoside is
almost equivalent to that observed for replacement of dJ with
dT (DDG = 11.35 kJmolÀ1). This is seen clearly when one
compares the binding constants of the sequence containing
dT (1i) [KD(1i) = 1370 Æ nM] with those containing Man (1b)
[KD(1b) = 1330 Æ 50 nm], All (1c) [KD(1c) = 1220 Æ 44 nm],
Gal (1d) [KD(1d) = 1240 Æ 44 nm], and Rib (1g) [KD(1g) =
1150 Æ 36 nm]. It is hard to rationalize how the major changes
in the orientation of the O-glycoside of 1b, 1c, 1d, and 1g
within the major groove (Figure 3) can correspond to such
minor changes in JBP1 affinity unless it is the case that JBP1
has an alternative low-affinity binding site on the tel-J*
oligonucleotides that is remote from the nucleotide J*. This
observation is new in the context of DNA binding by JBP1
and one that warrants further study.
A question unanswered in previous studies of JBP1/dJ
binding is the nature of the interaction between the 5’ (J-1)
nucleotide and JBP1. As described above, Sabatini et al.[13]
showed that this interaction is nucleotide independent. What
our study reveals is that the critical molecular component of
the J-1 nucleotide interaction seems to be the pro-R non-
bridging phosphoryl oxygen atom. This phosphoryl group
locks the pyranosyl group into the edge-on conformation
necessary for optimal JBP1 binding.
Our study also has two reasonable implications from a
drug-design perspective. First, there may be a site within the
DNA-binding domain of JBP1 that accepts the edge-on
conformation of the b-d-glucopyranosyl ring complexed with
the DNA backbone phosphoryl atom. From the viewpoint of
molecular recognition this locus should be a slot or crevice in
the protein orthogonal to the plane of DNA binding. Second,
disruption of the edge-on conformation of the glucose moiety
in dJ (1a) at the genomic level, by disengaging the hydrogen-
bonding network should also prevent JBP1 binding. Small
molecules that bind to DNA and, in doing so, modulate DNA
structure, such as major- or minor-groove binders or inter-
calators, may all promote this effect. An X-ray crystal study
by Wang and co-workers[25] of a b-d-glucosylated DNA
hexamer complexed with the DNA intercalator daunorubicin
hints that this approach may be effective. In this DNA–
daunorobicin complex, the glucoside does not adopt the edge-
on conformation we have shown to be important for optimal
JBP1 binding. More recently, Sabatini et al.[13] have shown
that the minor-groove binder chromomycin A3[26] blocks JBP1
binding to telomeric dJ-containing DNA, although it is not
clear whether this effect arises from direct inhibition of the
minor-groove JBP1 nucleotide binding or indirect inhibition
by a pertubation of the edge-on conformation of the O-
glycoside within the major groove by chromomycin A3.
Further studies are warranted to answer this question.
In conclusion, a comparison of equilibrium binding
constants and MD simulations of a panel of telomeric ds-
DNA sequences containing HMdU O-glycosides suggests
that the protozoan DNA-binding protein JBP1 binds to a
preferred conformation of b-d-glucopyranoside within the
major groove of dJ-containing oligonucleotides. This con-
Figure 3. Images of the MD snapshots of the minimum-energy, most
populated conformations of the HMdU O-glycosides 1a–g within the
major groove of tel-J* ds-DNA.
1 phosphate in the DNA backbone but at the expense of the
equatorial hydroxy groups on C2 and C3. In Rib (1g) the ring
pucker of the furanose ring in the major groove corresponds
to a C2 endo envelope conformation, with the C2 hydroxy
group forming a hydrogen bond with the pro-R phosphoryl
bond of the J-1 nucleotide.
When the MD simulations of the tel-J* oligonucleotides
are compared to their respective JBP1 binding affinities
(Table 1), it becomes clear that while JBP1 binds to all the tel-
J* sequences 1a–i, it does so with the highest affinity when the
HMdU O-linked glycoside can adopt a conformation that is
locked by hydrogen bonds, established between the hydroxy
groups of C2 and C3 of the pyranosyl ring and the pro-R
phosphoryl oxygen of the J-1 nucleotide. This H-bond
network and locked conformation of the b-d-pyranosyl ring
is observed with native Glc (1a), 4dGlc (1e), and Xyl (1 f)
(Figure 3). When the pyranoside can adopt this H-bond-
locked conformation within the major groove, the binding
data with Xyl (1 f) and 4dGlc 1e reveals that deletion of the
C5 hydroxymethyl group of 1a costs 3.61 kJmolÀ1 in binding
energy (equivalent to one typical-strength hydrogen bond) to
JBP1, and deletion of the C4 hydroxy group of 1a costs
7.39 kJmolÀ1 of binding energy (equivalent to one to two
hydrogen bonds) to JBP1. This data suggests that the hydroxy
groups of C4 and C6 of 1a are bound by groups within JBP1
2842 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 2839 –2843

Angewandte
Chemie
[7] A. Bernards, N. van Harten-Loosbroek, P. Borst, Nucleic Acids
formation is established by hydrogen bonds between the
equatorial C2 and C3 hydroxy groups of the sugar and the
pro-R phosphoryl oxygen of the J-1 nucleotide. This obser-
vation has clear ramifications for drug design against eukary-
otic parasites that are a truly global menace.
Res. 1984, 12, 4153.
[8] E. Pays, M. F. Delauw, M. Laurent, M. Steinert, Nucleic Acids
Res. 1984, 12, 5235.
[9] F. van Leeuwen, R. Kieft, M. Cross, P. Borst, Mol. Cell. Biol.
1998, 18, 5643.
[10] M. Cross, R. Kieft, R. Sabatini, M. Wilm, M. de Kort, G. A.
van der Marel, J. H. van Boom, F. van Leeuwen, P. Borst,
EMBO J. 1999, 18, 6573.
[11] P.-A. Genest, B. t. Riet, C. Dumas, B. Papadopoulou,
H. G. A. M. van Luenen, P. Borst, Nucleic Acids Res. 2005, 33,
1699.
[12] R. Sabatini, N. Meeuwenoord, J. H. van Boom, P. Borst, J. Biol.
Chem. 2002, 277, 958.
[13] R. Sabatini, N. Meeuwenoord, J. H. van Boom, P. Borst, J. Biol.
Chem. 2002, 277, 28150.
[14] M. de Kort, E. Ebrahimi, E. R. Wijsman, G. A. van der Marel,
J. H. van Boom, Eur. J. Org. Chem. 1999, 2337.
[15] J. J. Turner, N. J. Meeuwenoord, A. Rood, P. Borst, G. A.
van der Marel, J. H. van Boom, Eur. J. Org. Chem. 2003, 3832.
[16] E. R. Wijsman, O. van den Berg, E. Kuyl-Yeheskiely, G. A.
van der Marel, J. H. van Boom, Recl. Trav. Chim. Pays-Bas 1994,
113, 337.
Experimental Section
The fluorescence anisotropy of the Cy5-labeled DNA duplex(20 n m)
in binding buffer (Hepes buffer (35 mm, pH 7.9), EDTA (1 mm), KCl
(50 mm), MgCl2 (5 mm) and DTT (1 mm)) was measured in an SLM
8100 spectrofluorimeter equipped with a thermostatted cuvette
holder (208C). The sample was excited at 590 nm and emission
measured at 680 nm, with correction for both the buffer background
and G-factor of the instrument. JBP1 was titrated into the sample, and
emission spectra (at the magic angle) and anisotropy were measured
after each addition. The emission spectra showed only marginal
changes in intensity and wavelength maximum on protein binding,
whereas the anisotropy changes were significant. The resulting
binding isotherms (anisotropy vs. JBP1 concentration) were fit to a
1:1 association model for anisotropy, according to Equation (19) in
the article by Bailey et al.[18]
[17] R. R. Schmidt, Angew. Chem. 1986, 25, 213; Angew. Chem. Int.
Ed. Engl. 1986, 25, 212.
[18] M. F. Bailey, E. J. C. van der Schans, D. P. Millar, J. Mol. Biol.
2004, 336, 673.
Received: November 14, 2006
Published online: February 13, 2007
[19] D. A. Case, T. E. Cheatham III, T. Darden, H. Gohlke, R. Luo,
K. M. Merz, Jr., A. Onufriev, C. Simmerling, B. Wang, R. Woods,
J. Comput. Chem. 2005, 26, 1668.
[20] W. D. Cornell, P. Cieplak, C. I. Bayly, I. R. Gould, K. M.
Merz, Jr., D. M. Ferguson, D. C. Spellmeyer, T. Fox, J. W.
Caldwell, P. A. Kollman, J. Am. Chem. Soc. 1995, 117, 5179.
[21] J. Wang, P. Cieplak, P. A. Kollman, J. Comput. Chem. 2000, 21,
1049.
[22] M. Basma, S. Sundara, D. Calgan, T. Venali, R. J. Woods, J.
Comput. Chem. 2001, 22, 1125.
[23] K. N. Kirschner, R. J. Woods, Proc. Natl. Acad. Sci. USA 2001,
98, 10541.
Keywords: DNA · glycosides · molecular dynamics ·
molecular recognition
.
[1] P. Borst, F. van Leeuwen, Mol. Biochem. Parasitol. 1997, 90, 1.
[2] J. H. Gommers-Ampt, F. Van Leeuwen, A. L. J. de Beer, J. F. G.
Vliegenthart, M. Dizdaroglu, J. A. Kowalak, P. F. Crain, P. Borst,
Cell 1993, 75, 1129.
[3] F. van Leeuwen, M. C. Taylor, A. Mondragon, H. Moreau, W.
Gibson, R. Kieft, P. Borst, Proc. Natl. Acad. Sci. USA1998, 95,
2366.
[4] D. Dooijes, I. Chaves, R. Kieft, A. Dirks-Mulder, W. Martin, P.
Borst, Nucleic Acids Res. 2000, 28, 3017.
[24] K. N. Kirschner, R. J. Woods, J. Phys. Chem. A 2001, 105, 4150.
[25] Y. G. Gao, H. Robinson, E. R. Wijsman, G. A. van der Marel,
J. H. van Boom, A. H. J. Wang, J. Am. Chem. Soc. 1997, 119,
1496.
[26] J. R. Lundblad, R. P. S. Kwok, M. E. Laurance, M. S. Huang, J. P.
Richards, R. G. Brennan, R. H. Goodman, J. Biol. Chem. 1998,
273, 19251.
[5] F. van Leeuwen, E. R. Wijsman, R. Kieft, G. A. van der Marel,
J. H. van Boom, P. Borst, Genes Dev. 1997, 11, 3232.
[6] F. van Leeuwen, E. R. Wijsman, E. Kuyl-Yeheskiely, G. A.
van der Marel, J. H. van Boom, P. Borst, Nucleic Acids Res.
1996, 24, 2476.
Angew. Chem. Int. Ed. 2007, 46, 2839 –2843
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2843